Rational patient in the internet era
When you or someone you care about get a diagnosis you want to fight the disease. You start reading about it on the internet. You find exciting new therapies, depressing research results, personal accounts, you find support groups, companies selling therapies, diagnostics, medical devices, special diet products. You quickly get overwhelmed. You talk to your doctor — but he is fed up with all this new information from unreliable sources, with patients who think they know better, with companies trying to sell quack medicine.
Should you try things on our own or should you just follow what doctors say?
Pro self research/experimenting/medication:
- It is a natural thing to do. There are big psychological benefits to doing it which can also feed back into better health results.
- In Inadequate Equilibria the author notes a lot of evidence that incentives in mainstream medicine are not perfectly aligned with the good of the patient and often you can do much better than what medicine offers. The linked book gives some advice on how to check if you are in such circumstances. Personally I have also some anecdotal evidence of it myself.
- There might be a big variance in how we react to various treatments and some patients do need individual therapies. The counter argument is that if there is a systematic way to reach the unique individual therapy — then that procedure should be the subject of standard medicine.
- Some experiments might be orders of magnitude cheaper if you can trust the subject. For example those that require strict diets — a formal setup to control what the subject eats the whole day is difficult — but if you experiment on yourself you can do it easily.
- There is lots of stuff on the internet to read. The doctors often don’t seem to be up to date even with official research, but there might be useful stuff even in what is not yet subject of randomized trials.
One well publicized example: https://en.wikipedia.org/wiki/Augusto,_Michaela,_and_Lorenzo_Odone
- There is lots of stuff on the internet which can be even actively harmful or just not work and kill your hopes. It is hard to filter it.
- There is really lots of stuff on the internet, and the internet machine tends to filter the most extreme, which are mostly just random flukes. This is Littlewood Law: there are more than 4 billion internet users now — things that normally would happen to you ones in a million years happen daily to someone on the internet. It is so extraordinary that when published on the internet it spreads and easily reaches us. But it has no meaning to our lives, it is a once in a million years thing. That means that it is rational to be conservative evaluating stuff from the internet— just like the mainstream medicine is.
- The incentives of official science might be misaligned — but what are the incentives of a random internet guy we don’t even know. This is again the Littlewood Law — there are people with motivations that break our imagination: The Lying Disease. Why Would Someone Want to Fake a Serious Illness on the Internet?, but there are also people with trivial commercial motivations and fake accounts. It is like a Gresham’s Law for information — the incentives for selling something with fake stories (or crazy compassion seekers) will always outdo the incentives for sharing your experiences with strangers. It used to be easy to find stories from fellow travellers — now you can not trust such information and Google just stopped linking to it.
For example: The Remarkable Thing That Actually Freed Me Of Back Pain — someone reports good results for a novel method of healing back pain. It might be astroturfing campaign, the author seems to be a health journalist so the incentives of her writing this article might be complex, but it does not look completely fake. Still the fact that there is someone who was helped by this therapy is not that informative — every therapy has someone who believes it improved his health. I think I would try it out if I had back pain and it was available — but a more systematic approach would be to make a list of all back pain therapies and evaluate them all. Some of them might be too costly to try, some might be dangerous — but it would be rational to try most of them probably.
One common theme in the chronic Lyme communities is that people who go into remission tend to attribute their response to whatever technique, supplement, or medication they were trying at the time.
from Hacker News discussion on a new Lyme cure. This is not just Lyme — it is very basic feature of our thinking, people have been doing this kind of inferencing for ever, and in a world of scarce information it used to be useful. By the way that discussion is a good illustration of the problems I am writing about here.
You can do all the reading on yourself and you can consult your doctor, but coordinating your research with similarly motivated people limiting wasted and duplicated efforts could be so much more efficient. Communities are also helpful with lifestyle advice, which is often outside of medicine, and can give emotional support. But there are also huge problems with them, they can amplify the irrationality inherent in each of us:
- A community built around a particular hypothesis/therapy will never admit that hypothesis was wrong.
- It is natural for people with similar ideas to gather together and build subgroups, with the problem above, even if the whole community is not limited to just one point of view.
- Emotional support might become more important than truth, generating cult-like behaviour with rationalizations against conflicting evidence.
https://slatestarcodex.com/2019/11/20/book-review-all-therapy-books/ — Scott Alexander on psychotherapy books, comprehensive as always from him, also very good comments.
Doctors are turning to YouTube to learn how to do surgical procedures, but there’s no quality control and related Hacker News discussion: https://news.ycombinator.com/item?id=21675158
Iron law of tradeoffs: “if it were a stupider idea no one would support it, if it were a better idea everyone would unanimously agree to do it.” — similar to Inadequate Equilibria linked above,
“people like him work for decades, quasi-retire into positions accepting grant applications, and have a conscious or unconscious agenda to preserve or validate their life’s work” — from a Hacker News discussion, also interesting comments on autoimmune diseases
Update: A good illustration of why science is conservative: https://medium.com/@jamesheathers/hurry-dont-rush-e1aee626e733